In the ever-evolving landscape of digital connectivity, businesses face a relentless demand for faster, more reliable networks. Whether it’s the surge in remote work, the proliferation of devices that can connect to the internet, or the exponential growth of data-intensive applications, the strain on network infrastructure is ever-growing.
In this era where every byte counts, network capacity planning emerges as a critical strategy for organizations seeking to stay ahead in the digital race.
But what is network capacity planning?
And how do you do it effectively?
This article provides definitions around network capacity planning, most common use cases, key metrics, top benefits, and best practices to keep in mind during the planning process and beyond.
Let’s dive in.
What is network capacity?
In simple terms, network capacity is the amount of data that can traverse a network at any given point.
When taking a broad view, it is an essential component of modern life. The demands placed on the infrastructure responsible for carrying data across the world have grown exponentially, with little signs of slowing down. At the start of the twenty-first century, less than half of all American households owned a personal computer. Today, that percentage has jumped to 95%.
Many essential applications and services are now cloud-based, which requires constant internet access, along with a proliferation of IoT (Internet of Things) devices, where everything from your Fitbit to your fridge now requires the ability to get and stay online. All of these things create an ever-increasing need for bandwidth.
To the individual company, network capacity is the bread-and-butter of daily operations. From the first instant message of the day, to the shared document with collaborators around the globe, to the remote meetings with clients and co-workers, network capacity is what keeps operations flowing smoothly. Inadequate capacity can lead to frustrating delays, dropped connections, and decreased productivity, all of which directly impact the bottom line.
Moreover, in today’s hyper-connected world, where customer expectations for seamless digital experiences are higher than ever, a faltering network can tarnish a company’s reputation and erode customer loyalty. This makes sufficient network capacity more than just a matter of convenience. It’s a fundamental business imperative that underpins competitiveness and growth in the digital age.
How is network capacity measured?
Network capacity is generally measured in terms of bandwidth, which is the amount of data per second that can be handled by a network.
Networks are often referred to by their maximum throughput. For instance, most residential connections in the US average around 255 Mbps (Megabits per second). If you’re lucky, you can get over 1 Gps (Gigabits per second).
If you connect via a traditional ISP, things like the quality of the physical networking cable, the type of cable connection (copper vs. fiber), and the amount of people also using the link can impact your bandwidth.
If you connect via satellite, your speed ranges anywhere from 25 to 150 Mbps, and things like inclement weather and your view of the sky can impact your connection.
Commercial links are often much larger, as the needs for multiple users accessing cloud-based services and applications, along with other key resources, tend to be much greater. It is not uncommon to see commercial links at 5, 10, or even higher gigabits per second.
What is network capacity planning?
If network capacity is simply the measurement of how much data can go across your network, then network capacity planning is forecasting how much data you will need to send now and in the future. This planning is based on a detailed, intelligent analysis of what your needs are, what trends they follow, what network capacity planning tools make sense to invest in, and what the current network capacity planning best practices are (we’ll touch on this later on).
5 use cases for network capacity planning
While this is an easy enough concept, actually calculating and planning for your future network capacity needs can be quite complex. It is, however, critically important to have a handle on this, as it can have a huge impact on your business.
Why?
Let’s look at five specific use cases.
1. Business growth
Apple computers famously started in Steve Jobs’ parents’ garage with two employees. Today, they have 528 offices with over 164,000 employees. As your own business grows and expands, so too will the amount of bandwidth needed for your employees and customers.
2. Seasonal or event-based traffic
Traffic spikes due to holidays, special events, or promotions are important to predict and plan for as much as possible to avoid outages or service disruptions when the need is greatest. However, you don’t want to waste money on over-buying capacity for periods of less usage if you can avoid it.
3. Cloud migration
The proliferation of cloud-based platforms has brought innovation and efficiency to many functions, but they have also brought increased need for reliable bandwidth to access them. Network capacity should always be a factor when considering moving to the cloud.
4. Performance optimization
Being able to adjust your bandwidth usage according to peak demand can save you time and money. For instance, scheduling a large data transfer job, such as moving a database backup file to redundant storage, to occur during off hours can help save your bandwidth for your busiest times.
5. Disaster recovery/redundancy planning
Maintaining business continuity in the face of a disaster is a huge challenge. Network capacity planning can help identify critical network paths and resources that require redundancy or failover mechanisms.
Key metrics of network capacity planning
You know you need a plan. But where should you start?
Network capacity planning starts with good network management, and investing in good network capacity planning tools. To see what your future needs will be, you need to know where you stand now and how much capacity you have.
Your first step should be getting rock-solid network monitoring tools in place to track important metrics, like the following.
Bandwidth utilization
You need to get a sense of when you use the most bandwidth, and when you use the least. This is crucial not only on a short-term basis (for instance, knowing your utilization is low from 2am to 4am EST might let you schedule large data transfers during that window) but also during longer ranges of weeks, months, or years.
This knowledge allows you to plan for things like maintenance and upgrades when it will have the least impact on your needs. It also lets you plan for busy seasons such as certain holidays for online retailers. For long-term trends, you want to look at the 95th percentile of bandwidth usage. Basically, you discard the top and bottom 5% of traffic as outliers, and what remains is a picture of your typical utilization.
Latency
This is the measure of the time it takes for a signal to get from its source to its destination. Usually measured in milliseconds, it can be affected by the distance it has to travel, the type of connection, and the capacity of the devices it has to pass through.
Jitter
Also measured in milliseconds, jitter is the variation in latency during a continuous data stream. It’s like latency’s annoying cousin. If your internet connection seems fine for normal browsing, but voice calls are choppy and audio/video keep dropping, that’s likely because of excessive jitter.
Throughput
Similar to latency, throughput is the measurement of the rate at which data is successfully transferred between the source and the destination. It is typically expressed as bits per second (or kilobits per second, megabits per second, or gigabits per second). Latency measures the delay or responsiveness of the network, while throughput measures the capacity or efficiency of the network in terms of data transfer rate.
CPU and memory utilization
It might be a little surprising to see this on the list, since we normally associate this with applications, but the routers and switches that handle your network transmissions have cpu and memory limitations just like your servers. It’s very important to make sure you’re not maxing out any of the devices your data has to cross.
Error rate
This is the number of corrupted bits, usually expressed as a percentage of the total sent. These errors can manifest in various forms, such as bit errors, packet loss, corrupted data, or transmission failures.
Traffic analysis
The last metric, but perhaps the most important, is knowing what the traffic on your network actually is.
There are several protocols across different vendors that export what is called flow data, including Netflow, SFlow, and J-Flow. These are enabled at the router and/or switch level and provide insight into which applications are generating the most traffic on your network and where that traffic is going to.
For instance, you might see that your internal internet connection is at 90% utilization during the hours of 2pm to 5pm, and then with the use of Netflow discover that most of that traffic is going to blizzard.com’s Diablo IV servers. That might lead you to suspect some people are bringing in unauthorized devices to play games at work.
Top benefits of network capacity planning
Adequately planning for network capacity brings numerous benefits.
Cost savings
Having enough bandwidth to cover your daily needs without overbuying is a fine line to walk. Adequately planning capacity will help you avoid overspending on excess capacity or emergency solutions in the case of an outage.
It will also save you in lost productivity. If your employees can’t get to the services they need to work, your entire roadmap could be put in jeopardy.
Optimized performance
Your employees, users, and customers will be able to get the most out of your services without worrying if things are slow or, even worse, unavailable. It also allows you to identify potentially under-used resources and re-allocate them where the need is greatest.
Reliability and availability
The best product or service in the world can’t compete if users can’t access it. And once you’ve earned the reputation for being unreliable, it’s very hard to shake it. When you have adequately planned for network capacity, you don’t worry about suffering critical network outages and developing a reputation for being unreliable.
4 best practices of network capacity planning
1. Understand business requirements
Is your organization pushing everything to the cloud? Is your product team working on a killer new service? Does your CEO have ambition goals for doubling your customer base? What kind of SLA (Service Level Agreements) are you offering customers and employees? Is your organization ramping up its investment in SaaS applications?
Knowing what’s on the roadmap for your organization is key to forecasting future capacity demands, and being able to scale to meet those needs.
2. Assess your current infrastructure
Can your current infrastructure meet your growing needs? Is your hardware approaching end-of-life? Do you still have vendor support? What does a hardware refresh look like? Do you have a test or lab environment, where you can experiment with new software or devices to make sure they meet your needs before putting them into production?
No device lasts forever, and it’s important to know where you stand and what your current infrastructure can handle when predicting future growth.
3. Monitor and Measure Performance
It’s crucial to have as much data as you can around what is and isn’t working now so you don’t make the same mistakes down the road. This also lets you create a proactive plan, so you have time to explore multiple options and determine the best fit for your organization. You don’t want to rush to get something in place as the clock ticks on important deadlines.
4. Review and update your plans regularly
Capacity should be a key discussion point for every new product, initiative, and service planned. You should also periodically check to make sure there haven’t been any changes. Maybe there are different, better options available that weren’t when the plans were first made.
Network capacity planning tools
A good NMS (Network Monitoring/Management System) is absolutely the most crucial tool you can have for network capacity planning.
Auvik’s Network Management is a vendor-agnostic, cloud-based network monitoring and management solution designed for both internal IT teams and managed service providers (MSPs). It supports automated network discovery and monitoring across over 15,000 devices from 700+ vendors using standard protocols like SNMP, NetFlow, sFlow, and more.
A key capability is Auvik TrafficInsights, which leverages machine learning to provide real-time traffic analysis beyond just NetFlow data. This enables deep visibility across encrypted web and application traffic. With TrafficInsights, you can see all traffic flows across enabled network devices including traffic by application, protocol, IP address, and port. You also get visual representations of where traffic is going once it leaves the network.
For example, you’ll quickly be able to see:
- Who’s using banned applications, such as BitTorrent
- Who’s hogging bandwidth and slowing down the network
- Why your client’s web server is receiving so many connections
- What a hacked server was connected to during an infection
- When network traffic is spiking throughout the day
All of this can be seen on customizable, user-friendly dashboards, which make tracking and spotting trends easy while avoiding information overload.
Speaking of visualizing data, Auvik also offers robust reporting, which equips you to spot trends around network traffic patterns, protocol performance, and more, so you can make data-based decisions around capacity growth and other investments.
Effective network capacity planning is essential for organizations to ensure optimal performance, reliability, and scalability of their network infrastructure. By accurately forecasting future traffic demands, identifying potential bottlenecks, and implementing proactive measures, businesses can avoid costly downtime, maximize resource utilization, and support their evolving technology needs.
The next best step is to make sure you have a rock solid NMS system in place (or an MSP that provides networking services). That way, you can watch your traffic monitoring take shape and make plans for a successful future.
